Swarming boids using NanoBrain

We have started working on some projects using a small designed neural network we call a NanoBrain. This is because we think various create behaviours match well to a neural network. The first test application of this is the boids algorithm.

Each of the separation, cohesion and alignment rules are implemented with a neuron and their effects are merged by the root neuron. Settings the weights for each connection changes the swarming behaviour.

In the picture above you can see the main part of the network while running. The colors of the neurons represent their state: red is inactive and grey values represent the output values of each neuron.

There is an additional neuron names 'Avoidance' which is used to keep all boids contained within a cubic space.